t cc tr ng d
ng ri ro
c Th
ng i hc Khoa hc T
Luc
: 60 46 15
ng dn: TS.Trn Tr
o v: 2011
Abstract:
(1943) v c trm v min hp dn ci,
Q--c tr.
ng dt cc tr ng r
ng ri ro.
Keywords: ng; R ; t; Th
hc
Content
Lung:
Chng 1: Tp trung l r c kt qu ch ca l thuyt cc tr.
qu cho phn ng dng chng 2. i cc tr, xp x
i vt ngc tr cho khi cc u kin c
mm trong min hp dn ci G.
1 . Chi ting Fisher and Tippet (1928), Gnedenko (1943))
c tr.
2 Min hp dn ca mt phn phi G.
3 H phn phi vt ngng
4 Phn phi Pareto tng qu
5 H phn v
6 Biu QQ-PP
7 c lng c m h cc tr
8 Mt s m h cc tr m rng v mi lien h c m hh
Chng 2: dng l thuyt cc tr trong o lng ri ro t ch.
1. Ri ro t ch
2. M h o lng ri ro
3. Tham s ho bin li nhun v bin thua l
4. Mt s phng ph t c ri ro
5. Phng ph t gi tr ri u t vn
6. ng dng l thuyt cc tr trong m h ho ui chui li sut chng
kho
7. dng o lng ri u t c phiu ACB.
I. C kt qu ch ca lun vn t c:
II.
ng r
i ro cht ch
2.2 ng ri ro
Trong qun tr ri nu ch n thun d
ng h ng
l c rn thi ro thc ch h
c chn ca kt qu i ta thng s d
lng ru ch ng r
Giá trị rủi ro (Value at Risk-c s d
bin trong qun tr ri ro th trng ca danh mc. My, VaR vng hn
ch phng dit ln thc tit s dng
Tổn thất kỳ vọng (Expected Shortfall-ng ri ro th trng.
2.2.2 i ro cht ch
2.2
u t c t chc) nm gi mt danh mc. t i
m hin ti, (t+1) m cui ca k u t (thm trong tng lai), V
t
, V
t+1
ca danh mc t
c th
m t, t+1 tng V
t
t, V
t+1
cha
bin ngm gi danh mu t s i mt vi ri ro:
u t s b thua l, tn tht nu V
t+1
< V
t
c thua l: X = V
t+1
- V
t
n
ngV
t th i ro), mt ch nh
lng va th hin m ri ro ca danh mc (mc thua l) bt k ngun gc
ng ca th trng, t t, v n va thun tin cho
n tr?
i ro cn phng nhu c b
hc tin?
a nha th k trt s
gi [10u vn xut mt mô hình lý thuyết về độ đo
rủi ro c g ri ro cht ch ng ri ro ca danh mc.
2.2 i ro cht ch
Hong ca th trng bng
t
( , , )P
. Gi X
0
n ngu
u hn (hu nh chc chX X
0
i.
u t tham gia th trc nm gi danh mc. R
ca vic nm gi danh mc biu hin bi mc thua l tim n sau k u t
i bin ngXX.
2.1:
: X
gĐộ đo rủi ro ca danh mc. Danh mc vi
mc thua l X c ri ro
(X). Trong qun tr xem
(X) nh
khon d n th chp,
i ro
(X) gĐộ đo rủi ro chặt chẽ nu thu ki)
sau:
T1: Bt binh tin:
Vi mi XX, a
:
(X+a) =
(Xa
T2: C di:
Vi mi X
1
, X
2
X :
(X
1
+X
2
)
(X
1
) +
(X
2
)
T3: Thun nht dng:
Vi mi XX
X
(X)
n
Vi X
1
, X
2
X X
1
X
2
(hu nh chc ch
(X
1
)
(X
2
)
gii t nh sau:
T1: Vi danh m ri ro
(X), khi b n phi r a
m ri ro ca danh mc gi
(Xa.
T2: Ri ro ca danh mc tng hp (ng vi X
1
+ X
2
n ln tng ri ro ca
anh mp v Đa dạng hóa đầu
tư.
T3: Danh mn.
T4: Danh mc thua l tim
Nh vy tt c i v u h p vi
thc tin.
i ro ca danh mp ct t
n tr r nguồn gốc của rủi ro th.
Kt qu sau s cung c i ro c th
ri ro cht ch
2.2u di
i ro cht ch
(pu gi
1
0
( ) 1x dx
.
xem (p) nh mt d t.
Gi F(xi ca mc tn tht X i ro nh sau:
1
1
0
( ) ( ) ( )X F p p dp
(2.1)
vi (p(p
1
0
( ) 1x dx
coi (p trng.
(xem [10]):
t ch khi (p)A.
Vc biu din (2.1) c i ro cht ch
th p vi ngun gc ch(p
gt quan trng ri ro.
2.3 ri ro (Value At Risk)
2.3.1 Ngun gn
Thut ng ri ro (Value at Risk
c s dng r
c s tr t
m quan trng trong khoa hc kinh t t sau s kin th
trng ch
Ngp c
la chn danh mu t
trn hip phng sai li su n phi u danh mu t.
Trong nhu thy ban Chi h
K (SEC - bu
cu v vn l tin cy 95% trong
khong th i li su c s dng
n l ting thn
khai s dng mt h th vu t c
Trong thi gian cui thu tht s t chc hin
h tr cho vi vu t n ch ri ro ca th trng.
Nhng s kiu nhy rt nhip
rc rn th di mc d kin honh m
trc. T khi tt c n vi tn tht,
u kin tt y ri ro ca hu h
ty. Ti ting - Dennis Weatherstone
i tng kt tn tht ca tt c
tra giao d
tn thn cho mc ng dng rt
t rt nhiu ph
p mi li c lng cn thi
l gii hn trong m
t
c lt phn c
t rng tt c t ph
nh lng v hoa h. Nh
bm c
ca h.
T chc t u t
n nh y hn na vic s dng VaR trong
qun tr ri t bi ng tn tht th tr
nhng tip cn tng t nh c s dng trong nhiu khoa
Hinh.
t trong nh
ng ri ro th trng cn,
danh mc. S d mm nhng tn tht
v m ca danh ma mn trong danh mc bi
u t c lng m tn thc hi ri ro.
2.3.2 m VaR
VaR ca danh mn th hin m tn th xi vi danh
mn trong mt chu k k n v thi gian) vi m tin cy nhnh.
2.1: Mu t quyu t mt khon tin lt danh mc c
phia r danh mu t m xung 50000USD.
Sau khi khn nhn st gim li nhun, anh ta mun bit
m tn tht t li ngay lp t mt
ht khon tiu t, nh lp vi thc t t
trng hp thit hi lm khi x ln ti
s ki c bi n tht t ng hp s
4000USD i m ca VaR.
Trong qun tr r s dng r ri ro
m tn tht danh mnh. Cho mt danh m
sung thnh n m ngng sao cho
tn tht danh mc trong khong thi gian nh
t
cho trc.
2.3
2.3.3.1 Tip c
Gi s rng mu t quynu t mt danh mn P. Ti thi
ca danh mu t
t
V
. Sau mt khong thi gian
t
, ti thi
m
k t t
ca danh mu t
k
V
.
tk
VVkV )(
cho
bit s ca danh mc P trong khong thi gian
t
.
()Vk
g -
P&L(k))
k
chu k ca danh mc.
n t:
- u t v th i vi P sau chu k k nt
P&L(k)
< 0) s b tn tht.
- u t v th i vi P sau chu k k nt
P&L(k) >
0) s b tn tht.
t
t
tt
: Biu di n sau khong thi gian
t
V
k
n ngn ngi F
k
t c
x
gPhân vị mức α F
k
. V
P&L(k) < 0 t u t trng v s b tn th x
Pr(P&L(k) x
) = 1 -
) = 1 -
u t n v s b tn tht.
V
t
V
k
: Biu din m
u t v th trng v, khi
0)( kV
tu t s chu tn
tht. P(
)(kV
u t chu tn tht di mc x
(x
Ngc lu t v th n v,
0)( kV
tu t s chu tn
tht. P(
)(kV
x
) = 1 - P(
)(kV
) = 1 - u t chu
mc tn thc x
(x
-
hai v th u tu t chu tn tht t
danh mc st gi hai trng hc cho nh
mc tn th
vy VaR ca mt danh mc vi chu k tin cy (1-
F
k
(x). i l
Nh v
)(kV
,
t u t
nm gi danh mc P sau mt chu k k, v tin cy (1- u t
tn tht mt khon s bng
VaR(k, α)
u kin hong.
2.2: : VaR(1
u USD. Nh vy vt 5%, trong m thng ca
u USD.
: Trong thc tn quc t:
Nu chu k c 5%.
Nu chu k
2.3.3.2 thit c
ri ro (VaR) ph thu mt s
php cn VaR phi tham s
ngi quy c thit ru t
ng vng b
tng quan vi nhau. Nc lng mi chui
th thit ca OLS b vi phm. Mt chuc gng nu k
vng, phhip phi theo th
t ca chui theo thi gian.
Bc ng : Mt bin
t
Y
t bc ng u
ttt
uYY
1
t
u
u trhng sai
p phng sai b
)()()()(
11
tttt
YEuEYEYE
.
vng ca
t
Y
i. Vi gi thii ta tin
r t thu .
tr n nht thit ph
Thi gian c nh: Gi thit khong thi gian
u khong thi gian. Chng hn, nu cho khong thi gian
mt tu m rng cho m
i chun: Trong mt s ph thit li su
si chun, ch tr mt s php cn VaR
phi tham s nh Monte Carlo.
2.3
t l
n,
:
& ( )
& ( )
t t t
t
P L k
r P L k rV
V
.
Do
t
V
n,
a li
t
r
.
2.3i sun
Gi thit chui li sut cn
t
r
i d chun. Vi gi thit
cn s dng hai tham s k vng (
lch chun (
) (hoc s
dc lng c VaR.
T gi thit
),(~
2
Nr
t
suy ra
)1,0(~ N
r
t
nh nh sau:
)(%)100)1(,1(
1
NngàyVaR
(2.2)
i chun.
2.3: u t nm gi mt khi lng c phi
m hin ti
V
t
= 100 tri ng, li su chun r
t
2
( , )N
vi = 3%.
ng, li su gi nh
0
. Vi m
= 5% VaR ca li sut :
-1,64 0,03 = -0,0492.
Suy ra VaR ca danh mc:
t
, 5%) = VaR
Li sut
t
= (- 0,0492)*100 = - 4,92.
Vy sau mt 5%, kh u t l u.
2.3c
Cho danh mc P: (w
1
, w
2
, , w
N
) vi li su n trong danh mc
n: r
i
2
( , )
i
N
vt :
i
1
w.
N
Pi
i
rr
;
i
1
w.
N
Pi
i
rr
;
2
W'.V.W
P
y li sut ca danh mc r
P
2
( , )
PP
Nr
. T ng t nh i vn
c VaR ca danh mc:
ppr
NngàyVaR
p
)(%)100)1(,1(
1
. (2.3)
: Nc P di d: P: x = (x
1
, x
2
N
) vi x
i
khon
tiu t n i
1
& ( )
N
ii
i
P L k rx
.
Vi gi thit li sun trong danh mc r
i
2
( , )
i
N
2
&&
& ( ) ( , )
P L P L
P L k N
2
&&
1
; '
N
P L i i P L
i
x r x Vx
.
- )100% ) =
P&L
+ N
-1
()
P&L
=
P&L
+ N
-1
(x)
1/2
(2.4)
Vi chu k i lng
P&L
c t b
th
- )100% ) = N
-1
(
1/2
(2.5)
3:
gi nh li sut danh mc (ho
si chu cn c lng hai tham s: k vng ( lch
chun () ( ma trn hip phng sai V v
c) c VaR, do
ph
. Trong thc t thit
vi phm
c l
.
ta
ng ph
i cho c lng VaR gn vi tn tht
trong thc t nht.
2.3.3.5 n ch c i ro VaR
VaR ca danh mn th hin m tn th xi vi danh
m n trong mt khong thi gian nhnh vi m tin cy nh nh. Tuy
, VaR
i
, VaR mi ch
mt tn l
VaR lt phn nh i (1% hay
ng xu-ng vi nhng din bin bt thng ca th trng), khi xy ra tn
tht, mc tn th d i ro ES d cho
li.
2.4 n tht k vng ES (Expected Shortfall)
Nh
, mc s dn bin trong qun tr ri ro th
trng, rng ca danh mng hn ch nhnh c
phng di t ln thc tin. M p cn m ng ri ro th
trng ca danh m c s dng th Tổn thất kỳ vọng (Expected
Shortfall ES).
s u v th i ro
phc lng .
2.4.1 m
Sa danh mi nhng trng hp tn
tht thc t ca danh mc vt ng vng) cc
tn thT
:
Tn tht k vng ca danh mc v tin cy (1- , α), i
lng k vu kin:
( ) ( / ( ))ES ES E X X VaR
. (2.6)
Nh mt s t u vit hn VaR, vic s d i ro ES th hin vi
lng r hg VaR.
2.4.2 Mt s t ca ES
ch
sau (xem [10]):
i ro cht ch ca danh mc.
M i ro cht ch a danh m biu din nh mt t
hp li ca ES v (X).
Nh vy via danh mc va thay th VaR
ng r hn va ch i ro u vit.
2.4.3 Phc nghim c lng ES
thuc lc
l X ca danh m i sut (loga li sut) ca danh mc nh bi:
( 1)
ln
t
t
V
r
V
ng t nh khi c lng VaR t s ling
c lng ES: ph .
Ph d nh v i ca li sut r: chng hn
i chun, T- Student, Pareto t, s li ca r,
s dc lng trong th lng (h
moment t c l c trng ca
h c lng ca VaR (xem [9]) ng ng (xem
[10]).
Ph a ra gi nh v i ca li sut r
c lng thc nghi
k thup x (phi suy, mng n c lng
(xem [11], [18], [21], [22]).
c c lng ES nh sau: Cho m
thng chn = 1% hoc 5%. Lp mc n: (X
1
, X
2
n
u X
i:n
th t th i ca mu, t
1:n
X
2:n
i:n
X
n:n
. Gi k n
a n
t p = n
- k. Nu n
p
mu c t t n k:
1: 2: :
:
n n k n
kn
X X X
X
k
(2.7)
c c lng thc nghit tham kho trong
[10]):
:
()
kn
VaR X
(2.8)
:
: 1:
( : )
()
(1 ) ( : )
kn
k n k n
X n nguyên
ES
p X pX n không nguyên
(2.9)
2.4: Trong , ta c lng thc nghim ES cho th trng ch
Vi hu a
c thu thp trong khong thi gian t n 6/2010 t ngun
VnDirect. c
m, ng
c (xem [10]): n =1116,
n
= 11,6
ra k
1
=11, k
2
p
1
=0,6; p
2
=0,8. S dc c lng (2.8), (2.9
c lng thc nghim ci sut th tr
VaR
VnIndex
(1%) = 4,604 (%); VaR
VnIndex
(5%) = 3,686 (%)
ES
VnIndex
(1%) = 4,731 (%); ES
VnIndex
(5%) = 4,249 (%)
T
t s nh sau mch t
Nu li sut th trng gii kh c gi
i 99% kh
ng xu, nu li sut th trng ging
c gim d mc gim d
i sut th tri chuy sau khi c
l s dc 2 theo th
n. Nu chu k ch) hoc 10
VaR
VnIndex_tun
(1%) = 10,294 (%); VaR
VnIndex_tun
(5%) = 8,242 (%)
ES
(1%) = 14,96 (%); ES
(5%) = 13,43 (%)
2.4.4
Tn tht k vng ca danh mc v tin cy (1- ).100%, n th
vt ngng
VaR
:
( / )ES E X X VaR
.
Mc tn tht k vng (ES) ca danh mc g xu i ro b
sung cho VaR nhm quan trng cn tr ri r
rDo cc tp h c lng ES c
phc bi cp ti danh mc tp nh
danh mc ca t chng.
References
Tài liệu tham khảo
1. Acerbi, C., Nordio, C., Sirtori, C.(2001), Expected Sho rtfall as a Tool for
Financial Risk Management, AbaxBank – Working Paper.
2. Danielsson , J.&de Vries, C. ( 1997), Tail index and quantile estimation with
wery high frequency data, Journal of Empirical Finance 4, 241-257.
3. Danniel De Waal (200 4), Statistics of Extremes- Theory and Applications,
John Wiley& Sons, Ltd.
4. Feter F. C hristoffersen (2003), Elements of Financial.Risk.Management.
5. Fotios C. Harmantzis, Linyan Miao, Yifan Chien, Empirical Study of Value-
at-Risk and Expected Shortfall Models with Heavy Tails, Working Paper -
Financial Analytics Group, Stevens Institute of Technology, August 2005.
6. Hull, J. and A. White, Value at Risk When Daily Cha nges in Market Vari-
ables Are Not Normally Distributed, Journal of Derivatives, (1998) 5.
7. Koji Inui, Masaaki Kijima, On the significance of expected shortfall as a
coherent risk measure, Journal of Banking & Finance 29 (2005) 853–864p
8. Longgin. M ( 2000), From value at risk to stress testing: The extreme value
approach, Journal of Banking and Finance 24, pp.1097-1130 .
64
9. Manfred Gilli, Evis Kellezi (2003), An Application of Extremme Value The-
ory for Measuring Risk.
10. McNeil. A. (1998), Calculating Qua ntile Risk Measures for Financial Re-
turn Series using Extreme Value Theory
11. McNeil. A. (1999), Extreme Value Theory for Risk Managers
12. Neftci, S., Value at Risk Calculations, Extreme Events, and Tail Estimation,
The Journal of Derivatives, (2000)
13. R
¨
udiger Frey, Alexander J. McNeil, VaR an d expected shortfall in portfo-
lios of dependent credit risks: Conceptual and practical insights, Journal of
Banking & Finance 26 (2002) 1317–1334p.
14. R D. Reiss & M. Thomas, Statistics Analysis of Extreme Values, with Ap-
plications to Insurance, Finance, Hydrology a nd Other Fields
15. Yasuhiro Yamai, Toshinao Yoshiba, Value-at-risk versus expected shortfall:
A practical perspective, Journal of Banking & Finance 29 (200 5) 997 –
1015p.
65